You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A model asset may have a different idea for defaults than the one in the plugin. There should be a way for it to propagate this.
The example here is llama.cpp lora scale. Lora weights come as assets. Their scale is 1 by default (plugin level), but there certainly can be cases where a model has better defaults.
The main problem is that assets are independent from plugins. So the asset can't possibly know what the loading plugin schema is.
Likely the best way to do this would be to add a dict to asset descriptions which is free to interpret by plugins however they see fit (and thus subject to convention, rather than implementation)
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
A model asset may have a different idea for defaults than the one in the plugin. There should be a way for it to propagate this.
The example here is llama.cpp lora scale. Lora weights come as assets. Their scale is 1 by default (plugin level), but there certainly can be cases where a model has better defaults.
The main problem is that assets are independent from plugins. So the asset can't possibly know what the loading plugin schema is.
Likely the best way to do this would be to add a dict to asset descriptions which is free to interpret by plugins however they see fit (and thus subject to convention, rather than implementation)
Beta Was this translation helpful? Give feedback.
All reactions